geometryGeo.h5 - object 'slantRangeDistance' doesn't exist

[Jeanine Engelbrecht]

Hi All,

I am trying to run the smallbaselineApp.py on a stack of geocoded  interferograms produced in SNAP. I load the (geocoded) DEM and local incidence angle files in the geometry datastes. When running the process, the script fails on the step "Correct_topography" with the following error:

open geometry file: geometryGeo.h5

read 2D incidenceAngle,slantRangeDistance from geometry file: geometryGeo.h5

Traceback (most recent call last):

..

..

..

File "h5py/h5o.pyx", line 190, in h5py.h5o.open

KeyError: "Unable to open object (object 'slantRangeDistance' doesn't exist)"

Inspecting my geometryGeo.h5 file with the info.py suggests that only the height and incidence angle is present, and not the slantRangeDistance parameter. 

The slantRangeDistance parameter is supposed to be determined internally in load_data.py phase by calling the function:

https://github.com/insarlab/MintPy/blob/master/mintpy/utils/utils0.py#L60

but it seems in this case it is not calculated automatically. Can anyone please provide some guidance on how this issue can be solved? I would value any guidance.

All the best,

Jeanine

[Yunjun Zhang]

Hi Jeanine,

Similar to the incidenceAngle, slantRangeDistance can also be generated using an individual script: range_distance.py in this case. As you can see here (https://github.com/insarlab/MintPy/blob/master/mintpy/utils/utils0.py#L26), it requires a few metadata as well. You could check the metadata with info.py.

I hope this helps,

Yunjun

[rino salman]

Hi Yunjun,

I have processed ALOS2 data using alos2App.py and now I want to use MintPy for doing time series analysis.

I have all the required metadata listed here https://mintpy.readthedocs.io/en/latest/api/attributes/, except incidence angle and slant range angle.

I have generated both incidence angle and slant range distance based on your pointers here. However, I just realized that we can also read incidence angle from *.los.geo, I am surprise that the values are different from the computed incidence angle (attachment1). I am not sure which I should use, because it is needed for tropospheric correction, so I think we need the most accurate one.

Also, do you happen to know whether we can also read slant range distance from alos2App.py output? So that I have a consistent incidence angle and slant range distance with my unwrap.geo file (attachment2).

Best,

Rino

[Yunjun Zhang]

Hi Rino,

For the incidence angle, you should use the one from *.los.geo file. incidence_angle.py should generate 2D pixel-wise incidence angle for radar coordinates only because it calls range_distance.py internally, which assumes all pixels in the same column shares the same slant range distance. Therefore, this also means the slantRangeDisance from range_disntace.py in your case is also not usable.

incidence_angle.py should output 2D constant incidence angle for files in geo coordinates. MintPy determines a file is in geo-coordinates if it has metadata "Y_FIRST"; otherwise, the script will identify the file as in radar coordinates. I guess this might be the reason why you got spatially variable results from both incidence_angle.py and range_distance.py.

I am not aware of existing functions/scripts in isce to produce slant range distance directly. But it is easy to do. Option 1: write 2D slant range distance in radar coordinate and geocode them, you need to write a few lines of code for the first part. Option 2: call mintpy.utils.utils0.incidence_angle2slant_range_distance(), as these two information are interchangeable.

You might want to reach out to Cunren Liang, as he is working on preparing alos2App for MintPy as well.

Yunjun

[rino salman]

Hi Yunjun,

Thank you for your reply and the pointers. I did the second option and got what I am looking for (attached).

Best,

Rino